Age | Commit message (Collapse) | Author | Files | Lines | |
---|---|---|---|---|---|
2023-11-08 | debug: Disable Unavailable tests.disable_unavailable | Tim Newsome | 1 | -5/+13 | |
They aren't always reliable. See #520. | |||||
2023-10-25 | Merge pull request #519 from lz-bro/fix_icount | Tim Newsome | 2 | -2/+9 | |
Support limits_icount_to_one in IcountTest | |||||
2023-10-24 | Support instruction count limit in IcountTest | liangzhen | 2 | -2/+9 | |
This is taking into account that the hardware limits count to 1. Signed-off-by: liangzhen <zhen.liang@spacemit.com> | |||||
2023-10-17 | Merge pull request #513 from lz-bro/nonexist_csr | Tim Newsome | 2 | -4/+8 | |
Make the non-existent csr configurable | |||||
2023-10-17 | Merge pull request #514 from lz-bro/timer_configurable | Tim Newsome | 4 | -7/+17 | |
Make CLINT address configurable | |||||
2023-10-17 | Merge pull request #515 from riscv-software-src/unavailable_halted2 | Tim Newsome | 1 | -0/+61 | |
debug: Add UnavailableHaltedTest | |||||
2023-10-17 | Make the non-existent csr configurable | liangzhen | 2 | -4/+8 | |
Signed-off-by: liangzhen <zhen.liang@spacemit.com> | |||||
2023-10-16 | Make CLINT address configurable | liangzhen | 4 | -7/+17 | |
Signed-off-by: liangzhen <zhen.liang@spacemit.com> | |||||
2023-10-13 | debug: Add UnavailableHaltedTest | Tim Newsome | 1 | -0/+61 | |
Test behavior when a hart becomes unavailable while halted. | |||||
2023-10-13 | Merge pull request #512 from riscv-software-src/pexpect | Tim Newsome | 1 | -0/+1 | |
debug: Document that pexpect is needed. | |||||
2023-10-11 | debug: Document that pexpect is needed. | Tim Newsome | 1 | -0/+1 | |
Resolves #510. | |||||
2023-10-11 | Merge pull request #503 from lz-bro/dis_timer | Tim Newsome | 2 | -0/+13 | |
Disable timer interrupt to fix some bugs | |||||
2023-10-10 | Disable timer interrupt to fix some bugs | liangzhen | 2 | -0/+13 | |
Signed-off-by: liangzhen <zhen.liang@spacemit.com> | |||||
2023-10-03 | Merge pull request #508 from riscv-software-src/set_available | Tim Newsome | 2 | -7/+30 | |
debug: Add Openocd.set_available() | |||||
2023-09-29 | Merge pull request #509 from riscv-software-src/interlock | Tim Newsome | 1 | -0/+1 | |
debug: Better interlock when interacting with gdb CLI. | |||||
2023-09-29 | debug: Add Openocd.set_available() | Tim Newsome | 2 | -7/+30 | |
This helper uses dmi_write commands to mark harts available/unavailable. | |||||
2023-09-29 | Merge pull request #507 from riscv-software-src/targets | Tim Newsome | 1 | -1/+2 | |
debug: Make Openocd.targets() tolerate blank lines. | |||||
2023-09-29 | Merge pull request #506 from riscv-software-src/interrupt_all | Tim Newsome | 1 | -3/+4 | |
debug: Fix interrupt_all() to restore state. | |||||
2023-09-29 | Merge pull request #505 from riscv-software-src/debug_hart | Tim Newsome | 1 | -3/+13 | |
debug: Add --hart command line option to gdbserver.py | |||||
2023-09-28 | debug: Better interlock when interacting with gdb CLI. | Tim Newsome | 1 | -0/+1 | |
Actually wait for the command to be echoed back. This means we won't be confused if there are extra newlines in gdb output. | |||||
2023-09-28 | debug: Make Openocd.targets() tolerate blank lines. | Tim Newsome | 1 | -1/+2 | |
2023-09-28 | debug: Fix interrupt_all() to restore state. | Tim Newsome | 1 | -3/+4 | |
2023-09-28 | debug: Add --hart command line option to gdbserver.py | Tim Newsome | 1 | -3/+13 | |
This lets you reproduce a test running on a specific hart. | |||||
2023-07-24 | Merge pull request #497 from riscv-software-src/unavailable | Tim Newsome | 2 | -10/+7 | |
debug: Re-enable unavailable tests, and fix them for github | |||||
2023-07-24 | debug: Tolerate more whitespace from OpenOCD CLI | Tim Newsome | 1 | -1/+1 | |
During the github workflow this character is \n, while on my computer it's ' '. I'm sure there's a good reason for that, but it doesn't seem worth figuring out what that reason is. | |||||
2023-07-24 | debug: Re-enable unavailable tests. | Tim Newsome | 1 | -9/+6 | |
2023-07-24 | Merge pull request #499 from riscv-software-src/debug_workflow | Tim Newsome | 1 | -0/+172 | |
debug: Actually run tests in github workflow. | |||||
2023-07-21 | debug: Actually run tests in github workflow. | Tim Newsome | 1 | -0/+172 | |
This should avoid problems like we just had where bad tests can break the OpenOCD workflow. These tests only run if any debug files are changed, so should have no impact at all on non-debug tests in this repo. This file is copied and then slightly changed from riscv-openocd. New changes are that cacheable steps (building spike, OpenOCD) are stored to the cache even if running the tests fails. | |||||
2023-07-20 | Merge pull request #496 from riscv-software-src/pylint_workflow | Tim Newsome | 2 | -5/+9 | |
debug: Only run pylint if debug files changed. | |||||
2023-07-19 | debug: Better comment the privilege tests. | Tim Newsome | 1 | -3/+4 | |
Just doing this to make a change in the debug files, which should now cause the pylint workflow to execute. | |||||
2023-07-19 | debug: Only run pylint if debug files changed. | Tim Newsome | 1 | -2/+5 | |
2023-07-18 | Merge pull request #493 from riscv-software-src/disable_unavailable | Tim Newsome | 1 | -6/+9 | |
debug: Disable unavailable tests. | |||||
2023-07-18 | debug: Disable unavailable tests. | Tim Newsome | 1 | -6/+9 | |
They have issues when run in a github workflow. | |||||
2023-07-17 | Merge pull request #489 from riscv-software-src/power_dance | Tim Newsome | 10 | -55/+196 | |
debug: Test OpenOCD behavior when harts become unavailable, using new spike mechanism | |||||
2023-07-17 | debug: Create UnavailableCycleTest | Tim Newsome | 1 | -0/+27 | |
Use new spike mechanism to test OpenOCD behavior when a hart becomes unavailable, and then available again. | |||||
2023-07-17 | debug: CeaseRunTest -> UnavailableRunTest | Tim Newsome | 2 | -3/+23 | |
Use new spike mechanism to test OpenOCD behavior when the current hart becomes unavailable while running. Create ThreadTerminated exception. | |||||
2023-07-17 | debug: CeaseMultiTest -> UnavailableMultiTest | Tim Newsome | 2 | -8/+30 | |
Use the new spike mechanism to test OpenOCD behavior when a hart becomes unavailable while running. Create CommandException. | |||||
2023-07-17 | Interact with OpenOCD CLI over stdin/stdout. | Tim Newsome | 1 | -39/+102 | |
It's a bit messy to read the log file to get the output, but it seems to be flushed often so that this works. Also, added the `targets` method for retrieving the list of targets, and `wait_until_running` method to wait until all targets are in a running state. | |||||
2023-07-17 | debug: Add support_unavailable_control property. | Tim Newsome | 8 | -0/+10 | |
2023-07-17 | parkOtherHarts() already defaults to cease | Tim Newsome | 1 | -1/+1 | |
2023-07-17 | Move "monitor targets" calls into a central place. | Tim Newsome | 2 | -3/+1 | |
2023-07-17 | Move `import random` | Tim Newsome | 1 | -1/+2 | |
Just so it's easier to quickly comment out code and hard-code the target to use without pylint complaining. This really should be a command line option. | |||||
2023-07-13 | Merge pull request #490 from riscv-software-src/flushregs | Tim Newsome | 1 | -4/+4 | |
debug: flushregs -> maintenance flush register-cache | |||||
2023-07-13 | debug: flushregs -> maintenance flush register-cache | Tim Newsome | 1 | -4/+4 | |
flushregs is deprecated. | |||||
2023-07-13 | Add an exclude list for known failing Hifive1 tests (#485) | Marek Vrbka | 1 | -0/+28 | |
* Add an exclude list for known failing Hifive1 tests This commit adds a list of known failing tests based on: https://github.com/riscv/riscv-openocd/issues/869#issue-1769176709 * Fix name of the HiFive1 flash target Signed-off-by: Marek Vrbka <133884222+MarekVCodasip@users.noreply.github.com> --------- Signed-off-by: Marek Vrbka <133884222+MarekVCodasip@users.noreply.github.com> | |||||
2023-07-10 | Merge pull request #486 from en-sc/en-sc/warning-repeat-read | Tim Newsome | 1 | -2/+15 | |
change warning check in RepeatReadTest | |||||
2023-07-07 | Merge pull request #487 from riscv-software-src/debug_path | Tim Newsome | 2 | -8/+4 | |
debug: Don't rely on RISCV env | |||||
2023-07-07 | change warning check in RepeatReadTest | Evgeniy Naydanov | 1 | -2/+15 | |
2023-07-05 | Merge pull request #484 from TommyMurphyTM1234/master | Tim Newsome | 1 | -2/+2 | |
Fix for test program compilation failures due to lack of `zicsr` extension in `-march=rvXX...` | |||||
2023-06-30 | debug: Don't rely on RISCV env | Tim Newsome | 2 | -8/+4 | |
That made sense when mostly people used riscv-tools, but now they get tools from all sorts of places and most of them are suitable for the debug tests. Also document RISCV_TESTS_DEBUG_GCC and RISCV_TESTS_DEBUG_GDB environment variables in the README. The github workflows that rely on these tests don't use the Makefile, but instead invoke gdbserver.py directly, so they're not affected by this change. Fixes #481 |